data exploration

genre

user

rating

try different methods for factorization

apply our model for factorization (implemented by R)

explore factors by linear regression

try to recover missing values

apply Latent factor model (funk-SVD) by fast implemented version

apply truncated SVD in sklearn

apply soft impute using fancyimpute package

compare prediction powers

explore embeddings in W factorization matrix

MDS

manifold methods

occupation

gender

age